Abstract: A white list inside or outside determining apparatus includes: a first feature data extracting unit which extracts first feature data from an image by using a first transformation formula created based on preliminary learning images; a second feature data extracting unit which extracts second feature data from an image by using a second transformation formula created from the preliminary learning images and application learning images; a first matching unit which performs matching between a registration image and a collation image by using the first transformation formula; and a second matching unit which performs matching between a registration image and a collation image by using the second transformation formula. Weights of a matching result of the first matching unit and a matching result of the second matching unit are changed according to the number of preliminary learning images and the number of application learning images.

Abstract: A face image registration device includes an image input unit that inputs face images of a subject person, and an others face retention unit that retains a plurality of others faces. The device further includes: a false alarm characteristic calculation unit that collates the face images of the subject person with the retained others faces, and calculates a false alarm characteristic of the face images of the subject person; a correct alarm characteristic calculation unit that collates the face images of the subject person with each other to calculate a correct alarm characteristic of the face images of the subject person; and a registration face image selection unit that selects a registration face image from the face images of the subject person by using the false alarm characteristic of the face images of the subject person and the correct alarm characteristic of the face images of the subject person.

Abstract: An object of the present invention is to provide an image search device and an image search method by which search efficiency is improved.

Abstract: A collating device includes a collation list, a collation unit and a comparison unit. The collation list is configured to retain a false alarm list including a registered image, a threshold value serving as a criterion for determining whether to perform alarm activation, and a false alarm person image. The collation unit is configured to collate an input image with the registered image and the false alarm person image managed by the collation list, thereby obtaining a similarity therebetween. The comparison unit is configured to compare: a larger one of a value of the similarity between the input image and the false alarm person image, which is obtained by the collation unit, and the threshold value; and the similarity between the input image and the registered image, which is obtained by the collation unit, thereby determining whether to perform the alarm activation.

Abstract: A face recognition device includes: an image capture unit, an area detection unit, a feature amount extraction unit, a distance estimation unit and a feature amount update unit. The image capture unit is configured to capture an image of an object. The area detection unit is configured to detect a given area of the object based on an image obtained as an image capture result of the image capture unit. The feature amount extraction unit is configured to extract a feature amount within the given area detected by the area detection unit. The distance estimation unit is configured to estimate a distance between the image capture unit and the given area. The feature amount update unit is configured to update the feature amount extracted by the feature amount extraction unit based on the distance estimated by the distance estimation unit.

Abstract: A meta classification section assigns the degree of accuracy capable of identifying an individual to person feature information containing a face, color, shape, etc., and person passage information and stores the metadata of the person feature information and the person passage information for each degree of accuracy, a discrete retrieval section retrieves the person to be retrieved in the high accuracy metadata stored by the meta classification section, a dependence determination section determines a person having a dependence on the person to be retrieved using a plurality of retrieval results hit in the discrete retrieval and the high accuracy metadata, a query generation section regenerates a retrieval key from the identifier of the person having a dependence and the identifier of the person to be retrieved, and a discrete candidate-to-candidate retrieval section retrieves the person corresponding to the regenerated retrieval key in the low accuracy metadata.

Abstract: An image search apparatus has: a dimension reducing unit for reducing dimensions of multidimensional feature data to be produced as approximate data; an approximate-data storing unit for storing the produced approximate data with corresponding to the multidimensional feature data before the dimension reduction; a search request receiving unit for receiving an identifier which identifies multidimensional feature data of a person to be searched, at least as a search key; an approximate-space searching unit for calculating distances between approximate data corresponding to the received search key, and plural approximate data stored in the approximate-data storing unit, and for arranging the distances in order of similarity; and a real-space final ranking unit for again performing a distance calculation with using the multidimensional feature data before the dimension reduction, on a group of results in which the similarity is high, and for determining final rankings.

Abstract: An image search apparatus has: a dimension reducing unit for reducing dimensions of multidimensional feature data to be produced as approximate data; an approximate-data storing unit for storing the produced approximate data with corresponding to the multidimensional feature data before the dimension reduction; a search request receiving unit for receiving an identifier which identifies multidimensional feature data of a person to be searched, at least as a search key; an approximate-space searching unit for calculating distances between approximate data corresponding to the received search key, and plural approximate data stored in the approximate-data storing unit, and for arranging the distances in order of similarity; and a real-space final ranking unit for again performing a distance calculation with using the multidimensional feature data before the dimension reduction, on a group of results in which the similarity is high, and for determining final rankings.

Abstract: An object of the present invention is to provide an image search device and an image search method by which search efficiency is improved.

Abstract: A challenge to be met by the present invention is to provide an analogous image retrieval result display apparatus and a method for displaying a result of retrieval of an analogous image that make it possible to simply, accurately ascertain a behavior history of a person to be retrieved in various camera environments. Person characteristic data (a face, a color, a figure, and the like) extracted from a plurality of cameras (11) are stored in a person characteristic database (14). A retrieval request receiving section (101) receives an identifier for identifying a person to be retrieved as at least a retrieval key, and a characteristic retrieval section (102) performs retrieval of persons matching the retrieval key from the person characteristic database (14) in descending order of similarity.

Abstract: A meta classification section assigns the degree of accuracy capable of identifying an individual to person feature information containing a face, color, shape, etc., and person passage information and stores the metadata of the person feature information and the person passage information for each degree of accuracy, a discrete retrieval section retrieves the person to be retrieved in the high accuracy metadata stored by the meta classification section, a dependence determination section determines a person having a dependence on the person to be retrieved using a plurality of retrieval results hit in the discrete retrieval and the high accuracy metadata, a query generation section regenerates a retrieval key from the identifier of the person having a dependence and the identifier of the person to be retrieved, and a discrete candidate-to-candidate retrieval section retrieves the person corresponding to the regenerated retrieval key in the low accuracy metadata.

Abstract: A multimedia data recording apparatus is disclosed that enables large-volume recorded data to be recorded for a long period in accordance with the importance of the data, by means of a simple configuration and simple processing, and without imposing a heavy load on the system. In this apparatus, a layer classification section classifies video data captured by a surveillance camera and input via a video processing section into a plurality of layers on a frame-by-frame basis according to importance. A data recording section assigns a file to each layer and records classified data in a data recording medium. If an information amount reduction section detects that the vacant capacity of the data recording medium has reached or fallen below a threshold value, it deletes a frame belonging to the lowest layer among the plurality of layers.

Abstract: An image search apparatus has: a dimension reducing unit for reducing dimensions of multidimensional feature data to be produced as approximate data; an approximate-data storing unit for storing the produced approximate data with corresponding to the multidimensional feature data before the dimension reduction; a search request receiving unit for receiving an identifier which identifies multidimensional feature data of a person to be searched, at least as a search key; an approximate-space searching unit for calculating distances between approximate data corresponding to the received search key, and plural approximate data stored in the approximate-data storing unit, and for arranging the distances in order of similarity; and a real-space final ranking unit for again performing a distance calculation with using the multidimensional feature data before the dimension reduction, on a group of results in which the similarity is high, and for determining final rankings.

Abstract: A surveillance system having surveillance terminals and a surveillance center is configured to perform effective hazard alert signaling, depending on the hazard type. The surveillance system comprises a plurality of surveillance terminals, which are connected to the surveillance center by a network, carries out hazard and alert signaling. One surveillance terminal detects a hazard and sends a hazard information signal reporting what hazard has just been detected to the surveillance center. The surveillance center receives the hazard information signal, identifies what type of hazard source and the surveillance terminal that sent the hazard information signal, selects surveillance terminals to be alerted to the hazard from among the plurality of surveillance terminals, depending on the identified type of the hazard source and its locality, and sends a hazard alert signal to the selected surveillance terminals.

Abstract: It is one objective of the present invention to provide an image processing device that can easily implement an image processing function consonant with the need or application of a user. The image processing device comprises task storage, a task manager, a component manager, and a software bus manager. The task storage controls task scenario information indicating components used for an image processing task and information as to couple the components, component information, and substances of the components. The task manager analyzes the task scenario information and determines components used for the task and a software bus connecting the components. The component manager generates, deletes, or controls the components. The software bus manager generates or deletes the software bus, and controls the data exchange across the software bus.

Abstract: A surveillance system having surveillance terminals and a surveillance center is configured to perform effective hazard alert signaling, depending on the hazard type. The surveillance system comprises a plurality of surveillance terminals, which are connected to the surveillance center by a network, carries out hazard and alert signaling. One surveillance terminal detects a hazard and sends a hazard information signal reporting what hazard has just been detected to the surveillance center. The surveillance center receives the hazard information signal, identifies what type of hazard source and the surveillance terminal that sent the hazard information signal, selects surveillance terminals to be alerted to the hazard from among the plurality of surveillance terminals, depending on the identified type of the hazard source and its locality, and sends a hazard alert signal to the selected surveillance terminals.

Abstract: A surveillance system having surveillance terminals and a surveillance center is configured to perform effective hazard alert signaling, depending on the hazard type. The surveillance system comprises a plurality of surveillance terminals, which are connected to the surveillance center by a network, carries out hazard and alert signaling. One surveillance terminal detects a hazard and sends a hazard information signal reporting what hazard has just been detected to the surveillance center. The surveillance center receives the hazard information signal, identifies what type of hazard source and the surveillance terminal that sent the hazard information signal, selects surveillance terminals to be alerted to the hazard from among the plurality of surveillance terminals, depending on the identified type of the hazard source and its locality, and sends a hazard alert signal to the selected surveillance terminals.

Abstract: A multimedia data recording apparatus is disclosed that enables large-volume recorded data to be recorded for a long period in accordance with the importance of the data, by means of a simple configuration and simple processing, and without imposing a heavy load on the system. In this apparatus, a layer classification section classifies video data captured by a surveillance camera and input via a video processing section into a plurality of layers on a frame-by-frame basis according to importance. A data recording section assigns a file to each layer and records classified data in a data recording medium. If an information amount reduction section detects that the vacant capacity of the data recording medium has reached or fallen below a threshold value, it deletes a frame belonging to the lowest layer among the plurality of layers.

Abstract: A system of detecting a motion vector for an object block in a current frame of an orthogonally transformed video with respect to a stored reference frame. The motion vector is detected by: extracting a predetermined pattern of lower frequency components from the object block; orthogonally inverse transforming the extracted lower frequency components within the pattern to yield the pattern of pixel data; searching the reference frame for a block having a maximum correlation with the pattern of pixel data; and calculating the motion vector from the object block to the block having a maximum correlation. As an orthogonal transform, DCT (discrete cosine transform) is preferable. Creating a row and a column image comprising sums obtained by adding all the pixels in each column and each row of the object block and applying the above searching and calculating process to the row and the column image enables the motion vector detection from a normal or noncompressed video.

Abstract: The present invention provides a surveillance system, surveillance terminals, and a surveillance center, the surveillance system being configured to perform effective hazard alert signaling, depending on the hazard type, and a hazard and alert signaling method. The surveillance system of the invention where a plurality of surveillance terminals are connected to the surveillance center by a network carries out hazard and alert signaling. One surveillance terminal detects a hazard and sends a hazard information signal reporting what hazard has just been detected to the surveillance center.